How do file updaters work?

[saitho]

Hi there,

I've been using the JS implementation and am currently looking into your Go implementation.
I want need to move some files and write the new version number to a VERSION file before the tag is created.

I guess that's what file updaters are for. I saw the example for NPM, but I did not see in the code where the changed files are committed to the repository?

[christophwitzko]

Hi @saitho, this is currently not really possible. The file updaters are for updating the version number after the release (tag). You can use the --dry flag in combination with the --version-file flag to get the expected release. This will create a .version-unreleased file containing the next release, but not create a version tag.

[saitho]

Hi @christophwitzko, too bad. :(

Are there any plans to support this in the future? I'd love to contribute!
Maybe you can add a page that compares the JS and Go implementation of semantic-release so I can get a better understanding on the use cases and goals.

[joshuabaird]

@christophwitzko Is this still the best workaround for writing a new file prior to the tag creation?

I'm thinking I could use the GH action to run with --dry and then do things with the new version by referencing steps.release.outputs.version in a future step. After this, I would re-run the action with dry:false to actually cut the release.